Palaeoclimates in central and south America since 18,000 BP based on pollen and lake-level records

Abstract

The last 18 ka of palaeoclimates are discussed at 3 ka intervals based on palaeoenvironmental changes shown in pollen and lake level records from Central and South America. A total of 61 pollen records fulfill the requirements of continuity and chronology to be used in this synthesis: 8 records from tropical and subtropical lowlands in Central and South America; 7 records from the Central American Highlands; 26 records from the High Andes, north and south of the equator; 2 records from the southern subtropics; and 18 from the southern Andes and subantarctic islands. Palaeoclimatic interpretation of the pollen records is semiquantitative, based on bioclimatic parameters such as e.g. elevational temperature gradients or latitudinal precipitation gradients that can be related to vegetation zonation. Additional palaeoclimatic information is obtained from lake-level records, derived either from dated shorelines, or from the limnological history of lake sediment sections. Large-scale palaeoclimatic comparison of the proxy data records indicates that full- and late-glacial climates throughout Central and South America were cooler than today, by about 4 to 5°C, but that moisture patterns showed regional differentiation. Thus, in both the southern subtropics and high latitudes in southern Chile, conditions were substantially moister than today during full- and late-glacial times, whereas everywhere else conditions were far drier than today. In terms of palaeocirculation, this scenario can be interpreted as a poleward shift of the subtropical as well as the westerly circulation, a scenario that has been suggested by palaeoclimatic simulations from general circulation models (GCM). Another major palaeoclimatic aspect predicted by the GCM simulations is the suppressed early Holocene summer monsoon in the southern subtropics. Although proxy records are few from this zone, their palaeoenvironmental history during that time shows a period of aridity that would support the model. One of the most controversial aspects of South American palaeoclimates concerns the existence of a Younger Dryas equivalent, a cold period between 11 and 10 ka BP. Based on the existing data from South America, that show a great diversity of palaeoclimatic trends, the concept of an interval that is characterized by the same climatic signal as documented for the North Atlantic seems too simplistic. Moreover, at this point, only very few records have been analysed in sufficient detail to yield unambiguous climatic information for the interval in question. The principal outcome of this palaeoenvironmental and palaeoclimatic synthesis is the better understanding of the type and quality of data available at present, and the realization of regions and periods where much more and concerted effort is needed to address problems of intra- and interhemispheric palaeoclimatic correlations. Only then can questions concerning the mechanism of climatic change be tackled successfully.